Differential expression of growth-associated protein (GAP-43) mRNA in rat primary sensory neurons after peripheral nerve lesion: a non-radioactive in situ hybridisation study.

An alkaline phosphatase-labelled anti-sense oligodeoxynucleotide probe specific for growth-associated protein messenger RNA (GAP-43 mRNA) was used for non-radioactive in situ hybridisation histochemistry to follow relative changes in GAP-43 mRNA content in lumbar primary sensory neurons (L4-6) after unilateral ligation of the sciatic nerve. In normal dorsal root ganglia (DRG) 16% of neurons expressed GAP-43 mRNA, and these cells belonged to a sub-group of intermediate-sized (32-50 microns diameter) and large (> 50 microns) neurons. The hybridisation signal detected in these cells was weak to moderate. One day after nerve ligature a significant increase in the number of GAP-43 mRNA expressing neurons in the ipsilateral DRG was detected involving particularly the very small (12-20 microns) cells, and small cell population (20-32 microns), though the hybridisation signal was less pronounced in this latter cell group. A significant increase in the cellular content of GAP-43 mRNA was detected in both cell groups when compared to the normal DRG by 2 days after the lesion. At later times (4, 7, and 10 days postinjury) the intermediate-sized and large cell subpopulations also showed an increase in the number of GAP-43 mRNA positive neurons, followed by a significant rise in their content of GAP-43 mRNA. However, they did not reach the same intensity of hybridisation signal as seen in the small and very small neurons. All DRG neurons showed a maximum of GAP-43 mRNA expression by 10 days postsurgery. At longer times there was a slight decrease in the content of GAP-43 mRNA towards 14 days postinjury, but mRNA levels remained elevated up to 28 days after nerve ligature, the longest time point examined in this study. The different onset and levels of GAP-43 gene expression in the rat primary sensory neurons after lesion of their peripheral branch axons further characterize the different subclasses of these cells and may reflect their different involvement in the plastic changes following peripheral nerve injury.

Production and characterization of an anti-peptide antibody specific for the growth-associated protein, GAP-43.

In an attempt to raise specific anti-GAP-43 antibodies, a C-terminal tetradecapeptide was synthesized based on the predicted rat GAP-43 amino acid sequence using the F-moc polyamide solid phase procedure. The synthesized carboxy-terminal peptide was purified by reverse phase HPLC, conjugated to bovine serum albumin (BSA) and used as an immunogen. Polyclonal antipeptide antibodies raised in rabbits were affinity purified and their specificity for GAP-43 tested by Western blotting. Brain and spinal cord homogenates and GAP-43 enriched synaptosomal membrane fractions were analysed either by SDS-PAGE or reverse phase HPLC. The anti-GAP-43 antibodies detected a major immunoreactive band at 43 kDa (B-50), and a minor immunoreactive band at 38 kDa (B-60) together with an additional protein-band at 68 kDa, which was related to the peptide carrier, BSA. Immunohistochemical studies using these carboxy-terminal antipeptide antibodies revealed a widespread distribution of GAP-43 immunoreactivity throughout the adult rat brain and spinal cord, in a pattern generally consistent with earlier histochemical studies. It is concluded that the C-terminal GAP-43 specific tetradecapeptide is a potent immunogen and provides a convenient tool for the production of sequence specific anti-GAP-43 antibodies.

Intraspinal cellular responses after spinal cord transection in rats: neuronal expression of growth-associated protein (GAP-43).

The methods of non-radioisotopic in situ hybridization and immunocytochemistry were used to visualize sites of GAP-43 expression after a mid-thoracic spinal cord transection in adult rats. Neurons which expressed moderate to high levels of GAP-43 mRNA and showed strong GAP-43-like immunoreactivity were located immediately above the lesion site as well as at greater distances from the lesion site in the lower cervical and mid-lumbar spinal cord. The results of this study suggest a widespread occurrence of lesion-induced neuroplastic changes and may indicate that the increase in GAP-43 expression can be caused by axotomy, deafferentation and increased compensatory motor activity in the spinal cord of paraplegic rats.

Functional consequences of modification of callosal connections by perinatal enucleation in rat visual cortex.

The effects of neonatal monocular enucleation (right eye) on the callosal connections in the rat visual cortex were studied by physiological and morphological methods. Evoked activity was recorded in the left hemisphere, i.e. contralaterally to the enucleated eye. After enucleation, trans-callosally evoked responses were recorded in a widened stripe of the lateral visual cortex. Compared with the controls, the responsive area was expanded laterally and medially, i.e. into the lateral part of the primary visual area and within the secondary visual cortex (lateral part). Within about 0.5 mm of the expansion, the responses did not differ from those recorded in areas with "normal" callosal connections. Morphological evidence is presented suggesting that this expansion of evoked responses with high amplitudes and short latencies corresponds to an extension of callosal connections with a high density of axon terminals in layers two and three. Further medially within the primary visual cortex, callosally evoked responses with low amplitudes and longer latencies were recorded. The main types of unit responses and characteristic interactions between visually and callosally evoked responses are shown and discussed. These results suggest that following neonatal enucleation (1) the callosal connections expand and form functional synapses in the lateral part of the visual cortex, (2) these connections can activate cortical neurons either directly or by mediation of associational connections between the lateral secondary and primary visual cortex areas and (3) callosal connections can interact with visually evoked potentials and unit responses.